T
Timothy Tsai
Researcher at Nvidia
Publications - 49
Citations - 3218
Timothy Tsai is an academic researcher from Nvidia. The author has contributed to research in topics: Fault injection & Fault tolerance. The author has an hindex of 24, co-authored 49 publications receiving 2632 citations. Previous affiliations of Timothy Tsai include Alcatel-Lucent & Bell Labs.
Papers
More filters
Journal ArticleDOI
Fault injection techniques and tools
TL;DR: This work uses hardware methods to evaluate low-level error detection and masking mechanisms, and software methods to test higher level mechanisms to evaluate the dependability of computer systems.
Proceedings ArticleDOI
Understanding error propagation in deep learning neural network (DNN) accelerators and applications
Guanpeng Li,Siva Kumar Sastry Hari,Michael J. Sullivan,Timothy Tsai,Karthik Pattabiraman,Joel Emer,Stephen W. Keckler +6 more
TL;DR: It is found that the error resilience of a DNN system depends on the data types, values, data reuses, and types of layers in the design, and two efficient protection techniques are proposed.
Proceedings Article
Transparent run-time defense against stack smashing attacks
TL;DR: Two new methods to detect and handle buffer overflow vulnerabilities in process stacks are presented that work with any existing pre-compiled executable and can be used transparently per-process as well as on a system-wide basis.
Proceedings ArticleDOI
SASSIFI: An architecture-level fault injection tool for GPU application resilience evaluation
TL;DR: This paper presents an error injection-based methodology and tool called SASSIFI to study the soft error resilience of massively parallel applications running on state-of-the-art NVIDIA GPUs.
Proceedings ArticleDOI
An approach towards benchmarking of fault-tolerant commercial systems
TL;DR: The benchmark shows that Prototype B suffers fewer catastrophic incidents than Prototype A under the same workload conditions and fault injection method, however Prototype B also suffers more performance degradation in the presence of faults, which might be an important concern for time-critical applications.